1. Field of the Invention
The present invention relates to an image processing apparatus and image processing method and, more particularly, to an image processing apparatus and image processing method for performing an image process to reconstruct the composition of a captured image.
2. Description of the Related Art
The spread of digital cameras boosts the opportunity to capture still and moving images on travels, at events, and the like. However, when the user views a captured image, he may have a different impression between a capturing target during capturing and an image after capturing (to be referred to as a captured image). Under these circumstances, demand has arisen for an editing function of dividing a captured image into a region where an important subject is captured (to be referred to as an important-subject region) and a background region, transforming the important-subject region, and reconstructing the composition to generate an image as if the image were captured again.
For example, according to the invention disclosed in Japanese Patent Laid-Open No. 2006-067521 (patent literature 1), the region of a captured image is divided, and a blur corresponding to the distance is applied to each region using separately-obtained distance information. According to the invention disclosed in Japanese Patent Laid-Open No. 2004-310686 (patent literature 2), a hemispherical 3D virtual space is created from a captured image, and the model of each region is arranged in the virtual space so that the composition in the captured image can be changed.
By applying the invention in patent literature 1 or 2, the degree of blur or composition of a captured image can be corrected. However, applying the technique in patent literature 2 requires accurately constructing the 3D space. It is difficult to assemble this technique into a device such as a digital camera.